Selected Univariate Sensitivity Analyses of Directly Observed HAART Relative to Self-Administered HAART.

NOTE: HAART, highly-active antiretroviral therapy; QALY, quality-adjusted life years. Each estimate based on 10 simulated randomized trials with 1000 women per trial. * Simulated through 0.75 log10 reduction in viral load in 65% of women, with 0.25 log10 response in the remainder. † Highest probability of vertical transmission incorporated upper-bound transmission probability for each maternal viral load, and lower-bound estimate for effectiveness of Caesarean section, while lowest probability incorporated lower-bound transmission probabilities and upper-bound estimate for effectiveness of Caesarean section. ‡ A health care intervention is “dominated” if it costs more, but provides less health benefit, than a competing intervention. A dominated health intervention is never preferred [50]. A health care intervention is considered to be “cost-saving” when it costs less a competing intervention; “highly cost-effective” when it costs less than the GDP per capita; and “cost-effective” when it is between one and three times a country's GDP per capita, given that the intervention provides more health benefit than a competing intervention [49], [50]. § Discounted to present value at 3% per annum. ¶ Incorporated upper- and lower-bound estimates for costs of highly-active antiretroviral therapy (HAART), peripartum zidovudine therapy, and delivery of directly observed HAART. ∥ Incorporated upper- and lower-bound estimates for costs of vaginal delivery and Caesarean section.